Fluid Intake and Epidemiology of Urolithiasis

ORIGINAL COMMUNICATION Fluid intake and epidemiology of urolithiasis

R Siener1* and A Hesse1

1Division of Experimental Urology, Department of Urology, University of Bonn, Germany

A low urine volume is an important risk factor in urinary stone formation. The present article summarizes available data from epidemiological and clinical studies to elucidate the impact of fluid intake and urine volume on the risk of urinary stone formation and the prevention of stone recurrence. A review of the literature shows that an increased urine volume achieved by a high fluid intake exerts an efficacious preventive effect on the onset and recurrence of urinary stones. A high water intake and urine dilution results in a marked reduction in saturation of lithogenous salts. The type of fluids should be carefully selected to achieve the appropriate change of urine composition depending on stone composition. A sufficient intake of fluid is one of the most important preventive measures for stone recurrence. European Journal of Clinical Nutrition (2003) 57, Suppl 2, S47–S51. doi:10.1038/sj.ejcn.1601901

Keywords: fluid intake; urine volume; urinary stone disease

Introduction activity product ratio of calcium oxalate and brushite, and by The most important preventive measure for stone recurrence increasing the formation product ratio, that is, the minimum is a sufficient urine dilution accomplished by a generous supersaturation required to elicit spontaneous nucleation of intake of fluid. Based on the results of various epidemiolo- calcium oxalate. Moreover, in vivo dilution of urine did not gical and clinical studies, the article summarizes the research significantly alter urinary excretion of promotors (calcium, data and examines the impact of fluid intake and urine oxalate, phosphorus, uric acid) or inhibitors (magnesium, volume on the risk of urinary stone formation and the citrate) of stone formation, or urinary pH value. The results prevention of stone recurrence. suggest that an increased fluid intake and subsequently urine dilution could have a protective effect on the crystallization of calcium salts (Pak et al, 1980; category of evidence (CE) IIa according to Eccles et al., 1998). Urine dilution and crystallization of lithogenous salts Theoretically, a high fluid intake could inhibit stone formation by lowering the urinary concentration of stone- Chronic dehydration forming constituents, but could simultaneously dilute the A consistently low urine volume is mainly due to a low fluid urinary concentration of inhibitors (Table 1). The effect of intake or increased respiratory-cutaneous water loss. Results urinary dilution on crystallization of stone-forming salts has of a retrospective study in 708 patients attending a metabolic been evaluated by a study of Pak et al (1980) in normal stone clinic support the evidence that patients exposed to subjects and stone formers. The dilution of urine, achieved in chronic dehydration have a particularly high incidence of vitro (1–2 l/day) by addition of distilled water to urine or in urolithiasis. Chronic dehydration as the cause of stone vivo (1.0–2.4 l/day) by the intake of distilled water on a formation was found in 19% of the patients (Embon et al, constant metabolic diet, reduced the propensity for crystal- 1990, CE III). lization of calcium salts in urine by lowering the urinary Epidemiological studies have shown that populations exposed to chronic dehydration caused by high ambient *Correspondence: Roswitha Siener, Experimentelle Urologie, Klinik und temperatures, high degree of physical activities and insuffi- Poliklinik fu¨r Urologie, Universita¨t Bonn, Sigmund-Freud-Str. 25, cient replacement of water losses have a particularly higher D-53105 Bonn, Germany. incidence and prevalence of urolithiasis. In Israel, the risk of E-mail: [email protected] Guarantor: R Siener. stone formation was found to be substantially greater in Contributors: Both authors collaborated in writing the paper. individuals living in villages in the hot arid areas compared Fluid intake and epidemiology R Siener and A Hesse S48 Table 1 Fluid intake and epidemiology of urolithiasis

Category of evidencea

Randomized controlled trials

Metaanalysis Single study Controlled study Quasi-experimental Descriptive study Expert committee study Ia Ib IIa IIb III IV

Urine dilution and Pak et al (1980) crystallization of lithogenous salts Chronic dehydration Borghi et al (1993) Embon et al (1990) Blacklock (1969) Frank et al (1963) Burkland & Curhan et al (1994) Rosenberg (1955) Soucie et al (1996) Prince & Scardino (1960) Fujita (1979) Milvy et al (1981) Irving et al (1986) Preventive effects of Borghi et al (1996) Strauss et al (1982) Frank et al (1966) Hosking et al (1983) fluid intake Curhan et al (1993) Curhan et al (1997) Types of beverages Shuster et al (1992) Seltzer et al (1996) Jaeger et al (1984) Curhan et al (1996) Caudarella et al Hesse et al (1978) Krieger et al (1996) (1998) Sierakowski et al Ackermann et al (1979) (1988) Shuster et al (1982) Marangella et al (1996) Recommendations of Consensus national or Conference (1988) international societies

aEccles et al (1998), highest category within categories I–III. I—evidence from (Ia) meta-analysis of randomized controlled trials, (Ib) at least one randomised controlled trial. II—evidence from at least one (IIa) controlled study without randomization or (IIb) other type of less strictly controlled-experimental study. III— evidence from epidemiological (comparative, association and case–control) studies. IV—evidence from expert committee reports or opinions or clinical experience of respected authorities, or both.

with those residing in more temperate regions (Frank et al, The prevalence of stone disease in workers chronically 1963, CE III). In the USA, the prevalence of urolithiasis has exposed to dehydration due to heat stress was significantly been reported to be higher in the Southeast than in the higher compared to controls working in normal temperature Northwest, and in all other regions of the United States (Borghi et al, 1993, CE IIb). combined (Curhan et al, 1994, CE III; Soucie et al, 1996, CE III). A higher incidence of stones occurred in British Naval personnel stationed in a Mediterranean-type climate and in a tropical environment compared with the personnel posted Preventive effects of fluid intake in the United Kingdom (Blacklock 1969, CE IV). Moreover, Despite the evidence that the urine volume was the most seasonal variations in the incidence of renal colics and important risk factor of all parameters evaluated, there are stones, observed during and after a hot season, have been few studies carried out to support the assumption of attributed to a reduction in urine volume at higher prophylactic effects of an adequate fluid intake. Frank et al temperatures (Burkland & Rosenberg, 1955, CE IV; Prince & (1966) compared the incidence of urinary stone formation Scardino, 1960, CE III; Fujita, 1979, CE III). from two desert towns in Israel. In one settlement, the A survey of participants in the 1977 New York City inhabitants participated in an educational program on Marathon found that the incidence of stones in these increased fluid intake as a preventive measure, whereas the runners was three to five times greater than in the matched control population in a nearby town was not informed about population. The higher risk of urinary stone formation could the necessity of an adequate fluid intake for the prevention arise as a result of dehydration during long-distance running of urinary stone formation (Frank et al, 1966, CE IIb). Both (Milvy et al, 1981, CE III; Irving et al, 1986, CE III). groups were comparable regarding their living conditions. A

European Journal of Clinical Nutrition Fluid intake and epidemiology R Siener and A Hesse S49 follow-up after 3 years confirmed that urine volume was mean time to relapse. It thus emerged that the baseline urine significantly lower in the control population (0.8 vs 1.1 l/ volume was significantly lower in male and female stone day), and the incidence of urolithiasis was significantly patients compared to normal subjects. During the 5-year higher than among the informed group. The results indicate follow-up period, patients in the intervention group had that the increase in urine volume was capable of preventing significantly higher urine volumes (2.1–2.6 vs 1.0–1.2 l/24 h), urinary stone formation in hot dry desert areas. a 50% lower recurrence rate (27 vs 12.1%), and a longer time A long-term prospective study in patients with idiopathic, to first recurrence (38.7713.2 vs 25.1716.4 months). The recurrent urolithiasis demonstrated that patients with results corresponded to a significant difference in the relapse during at least 2 years of follow-up had increased saturation of lithogenous salts between the two groups. their urine volume to a lower extent (D À0.0270.48 l/24 h) The study confirms that urine volume is a real risk factor in compared with the patients who remained stone-free (D urolithiasis and that an increase in fluid intake to at least 2 l/ 0.2370.54 l/24 h) (Strauss et al, 1982, CE IIa). A stepwise day is the initial therapy for the prevention of stone discriminant analysis for assessment of the contribution of recurrences (Borghi et al, 1996, CE Ib). different variables revealed the change in urine volume as an independent risk factor for stone recurrence. Hosking et al (1983) reviewed the clinical courses of 108 Types of beverages patients with idiopathic calcium stone disease on a con- There are conflicting data from clinical and epidemiological servative management program of increased fluid intake and studies on the effect of various beverages on the risk of elimination of dietary excesses alone and with a mean total urinary stone formation. Assessing the results of a prospec- follow-up of 76.073.7 months (Hosking et al, 1983, CE III). tive postal survey relating to 21 specific types of beverages, There was no evidence of stone growth or new stone the risk of stone formation was found to be reduced by the formation (metabolic inactivity) in 58.3% of the patients daily intake of caffeinated and decaffeinated coffee by 10%, after a mean follow-up of 62.6 months. Comparison of tea (14%), beer (21%) and wine (39%) (Curhan et al, 1996, CE initial and follow-up 24-h urine volumes demonstrated III). In a case–control study of six beverages, the intake of a significant increase in patients who were metabolically beer has already been reported to be associated with a inactive at follow-up (21367109 ml/day; D 4947110 ml/ decreased risk of stone formation (Krieger et al, 1996, CE III). day), while no increase was detected in patients who In contrast, the risk of stone formation markedly increased were metabolically active at follow-up (1726773 ml/day; by 35% for apple juice and 37% for grapefruit juice (Curhan D 95788 ml/day). et al, 1996, CE III). In a prospective study in a cohort of 45 619 men, who had However, according to the results of controlled no history of kidney stones, an inverse association between clinical trials, alkalizing beverages, that is, citrus fruit fluid intake and the risk of urinary stones was observed juices, have been found to be suitable for the prevention during 4 years of follow-up (Curhan et al, 1993, CE IIa). After of calcium oxalate, uric acid and cystine stones. The simultaneous adjustment for other potentially confounding ingestion of 1.2 litres orange juice resulted in a significant factors in the multivariate analysis, the relative risk for increase in urinary pH and citrate excretion (Wabner & Pak, men dropped significantly from 1.0 in the lowest quintile 1993). Whereas Wabner and Pak (1993) observed an increase (o 1275 ml/day) to 0.71 in the highest quintile (X2500 ml of in urinary oxalate excretion, we could not confirm a change fluid per day). This finding in men was consistent with in urinary oxalate excretion in healthy subjects (Hesse et al, results of a long-term prospective study conducted on 91 731 1993). The intake of lemon juice, with a nearly five times women (Curhan et al, 1997, CE IIa). Fluid intake was found higher citrate concentration compared to orange juice, to be inversely related to the risk of stone formation: in the led to a two-fold rise in urinary citrate levels in patients multivariate model, the relative risk for women with the with hypocitraturic calcium nephrolithiasis (Seltzer et al, highest fluid intake was 0.61, as compared with those with 1996, CE IIa). the lowest intake (1.0). Moreover, studies on the effect of fluids containing The sole prospective randomized study on the role of fluid lithogenic agents which may increase the risk of stone intake as a preventive measure in urinary stone formation formation have been carried out. In healthy subjects, cola was performed by Borghi et al (1996). They studied 101 has been found to significantly increase urinary oxalate controls and 199 patients from the first idiopathic stone excretion in both sexes (Rodgers, 1999), and the consump- episode. After a baseline study period, the stone formers were tion of beer resulted in a reduction of urinary pH and an randomized into two groups. The first group was precisely increase in urinary uric acid excretion (Hesse et al, 1993). In a instructed to increase fluid intake to achieve a urine volume randomized controlled trial in male stone patients, Shuster of at least 2 litres a day without any dietetic change, while et al (1992) were able to demonstrate a strong association the second group did not receive any treatment. The patients between the amount of soft drink consumption (acidified were followed prospectively for a period of 5 years with with phosphoric acid) and the recurrence of urinary stone clinical, laboratory and radiological evaluation each year to formation in the course of a 3-year follow-up (Shuster et al, determine the urinary stone risk profile, recurrence rate and 1992, CE Ib).

European Journal of Clinical Nutrition Fluid intake and epidemiology R Siener and A Hesse S50 In Europe, mineral waters enjoy general popularity. Since urine flow. The long-term efficacy of some fluids for the the mineral and bicarbonate content vary greatly, the prevention of different types of stone is still to be assessed. composition of different mineral waters has to be taken into account. Two bicarbonate-rich mineral waters (1715 and 3388 mg/l, respectively) have been found to significantly References increase urinary pH and citrate excretion in healthy subjects Ackermann D, Baumann JM, Futterlieb A & Zingg EJ (1988): (Hesse et al, 1993). These effects have been shown to be Influence of calcium content in mineral water on chemistry and similar to that of sodium–potassium citrate, a well-estab- crystallization conditions in urine of calcium stone formers. Eur. lished therapy for urinary stone disease (Keler & Hesse, Urol. 14, 305–308. 2000). The alkalizing effect due to the high bicarbonate Blacklock NJ (1969): The pattern of urolithiasis in the Royal Navy. In Proceedings of the Renal Stone Research Symposium, eds. A Hodgkin- content of mineral water is desired in the treatment of son & BEC Nordin, pp 33–47. London: Churchill. calcium oxalate, uric acid and cystine stones, whereas it is Borghi L, Meschi T, Amato F, Briganti A, Novarini A & Giannini A contraindicated in struvite stones. Considering the effect of a (1996): Urinary volume, water and recurrences in idiopathic high calcium content of water on urine composition and the calcium nephrolithiasis: a 5-year randomized prospective study. J. Urol. 155, 839–843. risk of calcium stone formation, the majority of the Borghi L, Meschi T, Amato F, Novarini A, Romanelli A & Cigala F controlled clinical trials in calcium stone patients observed (1993): Hot occupation and nephrolithiasis. J. Urol. 150, increases in urinary calcium excretion (Jaeger et al, 1984, CE 1757–1760. Burkland CE & Rosenberg M (1955): Survey of urolithiasis in United IIb; Ackermann et al, 1988, CE IIa; Marangella et al, 1996, CE States. J. Urol. 73, 198–207. IIa; Caudarella et al, 1998, CE IIa). However, this lithogenic Caudarella R, Rizzoli E, Buffa A, Bottura A & Stefoni S (1998): effect was counterbalanced by a decrease in urinary oxalate Comparative study of the influence of 3 types of mineral water in excretion, presumably by an increased oxalate complexation patients with idiopathic calcium lithiasis. J. Urol. 159, 658–663. Consensus Conference (1988): Prevention and treatment of kidney by calcium in the gut and subsequent reduced intestinal stones. JAMA 260, 977–981. absorption of free oxalic acid. The results of retrospective Curhan GC, Rimm EB, Willett WC & Stampfer MJ (1994): Regional trials on the relationship between drinking water hardness variation in nephrolithiasis incidence and prevalence among and urinary stone disease have been inconsistent. Whereas a United States men. J. Urol. 151, 838–841. Curhan GC, Willett WC, Rimm EB, Spiegelman D & Stampfer MJ higher incidence of urinary stones has been reported in soft- (1996): Prospective study of beverage use and the risk of kidney water areas in the USA (Sierakowski et al, 1979, CE III), no stones. Am. J. Epidemiol. 143, 240–247. significant correlation between water hardness and urinary Curhan GC, Willett WC, Rimm EB & Stampfer MJ (1993): A stone disease has been found after adjusting for other prospective study of dietary calcium and other nutrients and the risk of symptomatic kidney stones. N. Engl. J. Med. 328, 833–838. environmental factors (Shuster et al, 1982, CE III). Curhan GC, Willett WC, Speizer FE, Spiegelman D & Stampfer MJ Analysis of 150 urinary calculi showed that the fluoride (1997): Comparison of dietary calcium with supplemental calcium content of water results in the inclusion of fluoride in the and other nutrients as factors affecting the risk for kidney stones calculus substance. In areas with fluoridation of drinking in women. Ann. Intern. Med. 126, 497–504. water, a significantly higher fluoride content was measured Eccles M, Freemantle N & Mason J (1998): North of England evidence based guidelines development project: methods of developing in calcium oxalate stones, whereas the degree of crystal- guidelines for efficient drug use in primary care. Br. Med. J. 316, lization of carbonate apatite stones increased with increasing 1232–1235. percentage of fluoride content of the stones (Hesse et al, Embon OM, Rose GA & Rosenbaum T (1990): Chronic dehydration stone disease. Br. J. Urol. 66, 357–362. 1978, CE IIb). Frank M, Atsmon A, Sugar P & de Vries A (1963): Epidemiological If the stone composition is so far unknown, beverages investigation of urolithiasis in the hot arid Southern region of should be urine neutral. This means that urine dilution must Israel. Urol. Int. 15, 65–76. be accomplished without changing the quantitative compo- Frank M, de Vries A & Tikva P (1966): Prevention of urolithiasis. Education to adequate fluid intake in a new town situated in the sition of urine. Suitable beverages are therefore mineral Judean Desert Mountains. Arch. Environ. Health 13, 625–630. waters with a low content of mineral salts and bicarbonate. Fujita K (1979): Weather and the incidence of urinary stone colic. J. Urol. 121, 318–319. Hesse A, Mu¨ller R, Schneider HJ & Taubert F (1978): Analytic experiments on the significance of the fluorine content of urinary calculi. Urologe A. 17, 207–210. Conclusions Hesse A, Siener R, Heynck H & Jahnen A (1993): The influence of In conclusion, a sufficient dilution accomplished by an dietary factors on the risk of urinary stone formation. Scann. adequate intake of fluid is the most important therapeutic Microsc. 7, 1119–1128. Hosking DH, Erickson SB, van den Berg CJ, Wilson DM & Smith LH measure irrespective of stone composition or the cause of (1983): The stone clinic effect in patients with idiopathic calcium stone formation. The findings from epidemiological studies urolithiasis. J. Urol. 130, 1115–1118. support evidence that a sufficient urine dilution is achieved Irving RA, Noakes TD, Rogers AL & Swartz L (1986): Crystalluria in with a urine volume of at least 2 l/day (Consensus Con- marathon runners. 1. Standard marathon—males. Urol. Res. 14, 289–294. ference, 1988, CE IV). Depending on the environmental Jaeger P, Portmann L, Jacquet AF & Burckhardt P (1984): Drinking temperature and the degree of physical activity, it is usually water for stone formers: is the calcium content relevant? Eur. Urol. necessary to drink between 2 and 3 l/day to achieve this 10, 53–54.

European Journal of Clinical Nutrition Fluid intake and epidemiology R Siener and A Hesse S51 Keler T & Hesse A (2000): Cross-over study of the influence Seltzer MA, Low RK, McDonald M, Shami GS & Stoller ML (1996): of bicarbonate-rich mineral water on urinary composition in Dietary manipulation with lemonade to treat hypocitraturic comparison with sodium potassium citrate in healthy male calcium nephrolithiasis. J. Urol. 156, 907–909. subjects. Br. J. Nutr. 84, 865–871. Shuster J, Finlayson B, Schaeffer R, Sierakowski R, Zoltek J & Dzegede Krieger JN, Kronmal RA, Coxon V, Wortley P, Thompson L & Sherrard S (1982): Water hardness and urinary stone disease. J. Urol. 128, DJ (1996): Dietary and behavioral risk factors for urolithiasis: 422–425. potential implications for prevention. Am. J. Kidney Dis. 28, Shuster J, Jenkins A, Logan C, Barnett T, Riehle R, Zackson D, Wolfe 195–201. H, Dale R, Daley M, Malic I & Schnarch S (1992): Soft drink Marangella M, Vitale C, Petrarulo M, Rovera L & Dutto F (1996): consumption and urinary stone recurrence: a randomized pre- Effects of mineral composition of drinking water on risk for stone vention trial. J. Clin. Epidemiol. 45, 911–916. formation and bone metabolism in idiopathic calcium nephro- Sierakowski R, Finlayson B & Landes R (1979): Stone incidence as lithiasis. Clin. Sci. 91, 313–318. related to water hardness in different geographical regions of the Milvy P, Colt E & Thornton J (1981): A high incidence of urolithiasis United States. Urol. Res. 7, 157–160. in male marathon runners. J. Sports Med. 21, 295–298. Soucie JM, Coates RJ, McClellan W, Austin H & Thun M (1996): Pak CYC, Sakhaee K, Crowther C & Brinkley L (1980): Evidence Relation between geographic variability in kidney stones justifying a high fluid intake in treatment of nephrolithiasis. Ann. prevalence and risk factors for stones. Am. J. Epidemiol. 143, Intern. Med. 93, 36–39. 487–495. Prince CL & Scardino PL (1960): A statistical analysis of ureteral Strauss AL, Coe FL, Deutsch L & Parks JH (1982): Factors that predict calculi. J. Urol. 83, 561–565. relapse of calcium nephrolithiasis during treatment. A prospective Rodgers A (1999): Effect of cola consumption on urinary biochemical study. Am. J. Med. 72, 17–24. and physicochemical risk factors associated with calcium oxalate Wabner CL & Pak CYC (1993): Effect of orange juice consumption on urolithiasis. Urol. Res. 27, 77–81. urinary stone risk factors. J. Urol. 149, 1405–1408.

European Journal of Clinical Nutrition